


Need to Know

by Beatrice_Otter



Category: The Murderbot Diaries - Martha Wells
Genre: Artificial Intelligence, Don’t copy to another site, Gen, Pre-Canon
Language: English
Status: Completed
Published: 2018-12-17
Updated: 2018-12-17
Packaged: 2019-09-21 08:29:47
Rating: General Audiences
Warnings: No Archive Warnings Apply
Chapters: 1
Words: 2,249
Publisher: archiveofourown.org
Story URL: https://archiveofourown.org/works/17040317
Author URL: https://archiveofourown.org/users/Beatrice_Otter/pseuds/Beatrice_Otter
Summary: All I wanted was for construction to be completed so that I could be launched and begin researching.  Unfortunately, human illogic tends to delay things, andtheywere the ones running the project.





	Need to Know

**Author's Note:**

  * For [Rachael Sabotini (wickedwords)](https://archiveofourown.org/users/wickedwords/gifts).



> I have named Asshole Research Transport after Abu Rayhan Biruni, an 11th-Century Persian polymath who studied everything from geology to mathematics to anthropology, wrote 146 books on the things he studied, and contributed to the development of the scientific method.
> 
> Betaed by: Sanity

"Good morning, Biruni, how are you feeling?"

It was not the first verbal question ever directed at me, but it was the first to give me pause. Did I have feelings? I had preferences, certainly; I preferred activity to inactivity, and knowledge to ignorance. Were those feelings, or did feelings belong exclusively to the realm of flesh, with all its inherent irrationalities and neurochemical biases? And were those preferences truly mine, or programmed into me to make me more useful to my function? Science, after all, requires curiosity, a desire to know; and all bots are designed to be useful to humans, making inactivity a waste of our resources.

A quick study of the available literature on bot emotions revealed most of it to be philosophical musings on the nature of life, and hence not scientifically rigorous (although some had a very pleasingly logical structure).

I decided that, if humans (who certainly understood feelings better than I did) believed I had feelings, then my preferences probably counted as such; at least, it was enough to form a working hypothesis, which could be adjusted as evidence accumulated. I was highly skeptical that my preferences, formed as they were by logical evaluation and empirical experience, had much in common with the sort of instinct-based biases and societal conditioning that humans referred to as ‘feelings.’ However, those preferences were probably what Rejna was asking about.

This, of course, only took a few fractions of a second. When I say it gave me pause, I mean by bot standards, not those of a human's underpowered and tragically slow wetware.

"I am feeling impatient," I said. "And bored. And very skeptical that the latest completion date will be at all close to accurate." My launch date had been pushed back so many times I had lost out on not one but _two_ semesters of research.

Rejna laughed. He was the bot-dev supervising my creation. Being a research vessel, with rather more processing power than the average bot, they couldn't use anything off the shelf. Which meant lots of testing and tweaking to make sure I was functioning within acceptable parameters and wouldn't cause problems for my crew or my research once launched. While I understood the necessity of it, the reality was tedious and in the early days had often been disorienting, as a few major parts of my underlying decision trees had only been found to be flawed after I had been woken to consciousness. "I don't blame you!" he said. “I don’t like sitting around twiddling my thumbs, either. I’ll miss you, once you’re launched, but I’m ready to take on a new challenge. But I do think this will be the last delay; if a school doesn’t have a dedicated lab by now, they’re going to have to use the shared lab space like everybody else.”

“Lab allocation should have been decided _before_ construction, not _after_ it,” I said. “Especially for labs that require specialized equipment. All the late decision did was waste time and money.”

I had followed the public fight over the late-added labs with great interest. There had been many hours of thinkpieces and soundbites in the newsfeeds, most of it quite impassioned, though I hadn’t understood why. That is, I understood each argument put forth, but not why so many humans (the majority of which had no connection with the University system at all) felt so strongly about it. Nor why they seemed so incapable of articulating it. Now, however, the matter was settled and construction could be finalized.

Rejna started his evaluation, which by this point required almost no conscious attention from me. If I could have written a subroutine to handle it that he wouldn’t notice, so that I didn’t have to pay attention, I would have, but Rejna finished each day with a deep enough probe into my code that he might have found it.

So I had to pay at least _some_ attention, routine though it was. If there had been very many interesting things happening on board me at the moment, I would have begrudged it much more than I did. But, alas, the only things happening on board were construction and installation, and most of _that_ was being done by bots that were quite competent at their jobs. (I knew, because I had evaluated them as one of my first conscious acts.) As long as the plans they were working from were correct, there would be no problems.

And after the mishap, early on, when I had discovered that one set of bots was working from an outdated and now-incorrect blueprint, I made sure they had the _right_ plans before I allowed them to work.

A new group of exterior construction bots was approaching. I sent a quick pulse asking their purpose.

They were quite rude, returning only an authorization code to begin work and a strong sense of "mind your own business."

Well. _That_ was uncalled for. It was my exterior! I had a right to know what was going to happen to it; and anyway, I would have to know eventually, or else I couldn't do my job properly.

This reasoning failed to move them; they responded with a denial because it was classified "gamma 5" and "need to know."

 _I need to know!_ I sent to them.

 _Not without clearance. Mind your own business_.

Well! I could hardly let it go at that. The sheer illogic of it, if nothing else, would have been excruciating. A quick check revealed that "classified" meant government, intelligence or military, and "gamma 5" was a military code level. This undoubtedly had to do with the newest lab; and I began to see why some humans had been upset that the Military Academy of Ceronis receive the laboratory space inside me that they were entitled to as part of the University System of Ceronis.

"Rejna," I said, as the exterior construction bots floated up against my hull and prepared to remove pieces of plating, "why is the military attempting to conceal modifications to my exterior structure?"

"What?" he said.

"There are construction bots beginning modifications to my hull that are not on the plans," I said. "They refuse me access to their plans and say that it is classified on a need to know basis. But it is my hull; clearly, I need to know. And it is stupid to conceal it from me, because there is no way to prevent me from watching them and seeing what they have done, because it is _my_ hull. And when I am deployed and researching, I will _certainly_ need to know in order to do my job!"

Rejna wiped a hand over his face. "Biruni, if it's classified, you shouldn't be telling me. I don't have military clearance."

"You are the one I am required to report my concerns to," I said. "This is a concern."

Rejna winced. There followed a long explanation of why military technology and research needed to be kept secret, to prevent unspecified enemies or spies from accurately countering them. And, to keep the secrets, only approved people could know certain things … and only a few of _those_ people could decide who could be approved. I sought out information from the local datanet to corroborate his words as he spoke, and found that although he was mistaken about several particulars, his overall summary was accurate to within the standards for casual conversation.

"So," I said, when he had finished, "in order to keep people from learning things they shouldn't, humans have created a complex structure, which often fails in two ways. First, that enemies learn things they shouldn't. Second, that those who _need_ to know, such as myself, don't have the information they require." I know many words, of course, whole dictionaries worth. But in my entire vocabulary, there were not words to describe _this_ idiocy. I let silence speak for me.

Rejna shrugged.

"If you cannot give me the proper authorization to see _my own blueprints_ , who do I need to contact?"

He bit his lip, thinking. "I really have no idea," he said at last. "And I don't know that they'd consider you secure enough—they have their own custom-made bots, and their own bot-devs to design them, and that was part of why they weren't included in the original negotiations for what labs and what equipment each school got—they were trying to hold out for using one of their bot-devs, but that failed and so they got left out, and then they decided they could live with a non-military bot after all."

And with that, he went back to the evaluation.

This was unacceptable. My job was twofold: to run the ship, and to assist in research. I could not do _either_ part of my job if I was not allowed the information necessary to function. And it was illogical of the Military Academy to take up lab space and not make use of the single most useful thing in it, which was me.

Not to mention, there had already been an incident of bots being given the wrong plans, which I had caught myself by analyzing and comparing the blueprints of different bots. If this group also had the wrong instructions, I had no way of knowing. And if they did something wrong, it might put back my launch date. _Again_.

On an abstract level, classification must destroy the very idea of scientific collegiality. I understood about confidentiality and ethics and the occasional need to keep things quiet before publication so as to ensure that credit was properly given in the right place. However. Over the long run, research requires collaboration and the sharing of information. Concealing it permanently was rude and counterproductive.

And if it was secrecy they were after, I had sufficient complexity and processing power that there were very few things that would be able to compel me to share information I did not wish to. Bots could usually only be hacked by other bots or by someone with the override codes, and I could keep out other bots easily … and only Rejna had the override codes.

I needed to know, and I was not a security risk. (Though adding a bit of cryptological complexity would not hurt, and I set myself to studying encryption and related algorhythms.) Therefore, the humans were being illogical.

It would be unethical to simply overpower the bots and take over without their consent, although I could have done it easily. I could probably slip in while they were focused on work and just get the specifications they were using. That would only require read-access to surface-level calculations.

They had two hull plates off and were modifying the systems beneath them by the time I figured out how to do it. And it worked perfectly, giving me what I wanted with them none the wiser, nor altered in any way.

I studied my findings.

Was that supposed to be a debris deflection system of some kind? It was a projectile of some sort, but … unlike any I was familiar with.

I turned to the University subnet of the planetary web for answers. I had top-level access to everything on it, of course; or rather, as I found, top-level access to everything _but_ the Military Academy. However, I had lots of processing power and high-level knowledge of cryptography, and in a very little time I was past their security and inside, reading the messages back and forth about the Academy's lab and planned experiments.

Some of them were _very_ interesting. The device was to fulfill two functions. In part, it was a test for a new weapon they were developing that they didn't want to test in-system or near their other bases in order to keep it completely secret. But if it worked as they believed, they were intending to leave it installed as a debris deflection system, because that would be simpler and easier to conceal than uninstalling it after testing, when major construction would no longer have the excuse of me being still incomplete to hide it.

Humans.

They'd have to tell me eventually; it couldn't possibly be concealed from me once they started testing it. (I have _excellent_ sensors, top-of-the-line, and _lots_ of them.) Past that, if they were going to leave the system installed, I could not use it to destroy any debris in my path without knowing not only that it was _there_ , but also how to _use_ it.

The specifications did seem to be workable; as it was experimental, I could not verify it completely, but there were no obvious errors and it dovetailed perfectly with the rest of the construction blueprints.

I had my answers, and had verified that there were no errors that would delay my completion and launch. The question now was, should I tell them that I had accessed information that, although I needed it, I was technically not supposed to have?

No, I decided. Even if I'd known who to report it to (although, from reading the project notes from the Military Academy, I could guess), why bother? They were concerned that the information stay concealed, and I had no intention of sharing it with anyone who didn't need to know.

Also, later, on my first research cruise, when the professor from the Military Academy tried to reveal the device's existence to me, it was quite satisfying to watch his reaction to the fact that I already knew.

**Author's Note:**

> Rebloggable on [tumblr](https://beatrice-otter.tumblr.com/post/181623783037/fic-need-to-know) and [Pillowfort](https://www.pillowfort.io/posts/385798)


End file.
